Predictors of Weight Change After Long-Term Treatment With Brexpiprazole in Schizophrenia
Background: We assessed effects contributing to weight gain during long-term treatment with brexpiprazole.
Methods: Data from an open-label, 52-week extension study of brexpiprazole in patients with schizophrenia were used. A criterion of >7% weight increase at any time after starting brexpiprazole was used, and contributing factors were assessed using a logistic regression with relevant variables. A model including three regions (US/Canada, Europe, Other [Asia/Latin America]) was used to assess the potential effect of regions. A stepwise approach assessed other potential contributing factors. Step 1 identified how best to include BMI in the model (continuous or categorical). Step 2 introduced several other terms to assess if they better explained the change in weight. Terms investigated included: baseline weight (continuous), age (continuous, categorical), sex, baseline PANSS Total score, region interaction with BMI.
Results: The analysis set comprised 815 patients. Change in weight depended on region (p=0.0347); BMI was best included as a categorical variable ( < 25; ≥25 to < 30; ≥30; p<0.0001). No other variable contributed with an effect (p<0.05). Regional differences were reflected in the odds ratios (OR) of patients experiencing >7% weight gain (Europe vs US/Canada OR=0.781, Other vs US/Canada OR=1.422) as was BMI category (BMI < 25 vs ≥30, OR=2.617; BMI ≥25 to < 30 vs ≥30, OR=1.583). No interaction was seen between ‘region’ and ‘BMI category’ factors.
Conclusion: Most possible contributing factors did not predict weight gain. However, patients with BMI < 25 at baseline were more likely to have clinically significant weight increase. Region and BMI were important for predicting weight gain.
This poster was presented at the 32nd annual Psych Congress, held Oct. 3-6, 2019, in San Diego, California.
